From mboxrd@z Thu Jan 1 00:00:00 1970 From: Greg KH Subject: Re: bcache oops in bch_insert_data with the latest stables fixes (3.10.15) Date: Fri, 11 Oct 2013 08:51:51 -0700 Message-ID: <20131011155151.GA17979@kroah.com> References: <52513DBD.80005@excellency.fr> <525186E3.4000905@gmail.com> <525483A4.5020607@gmail.com> <20131008223221.GD7202@kmo> <20131009212552.GA24271@kroah.com> <20131011070736.GA11822@kmo-pixel> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <20131011070736.GA11822@kmo-pixel> Sender: linux-bcache-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Kent Overstreet Cc: Gabriel de Perthuis , linux stable , cbay-SxHCd5+OuqTrt3ojHgZu+w@public.gmane.org, "linux-bcache-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" List-Id: linux-bcache@vger.kernel.org On Fri, Oct 11, 2013 at 12:07:36AM -0700, Kent Overstreet wrote: > On Wed, Oct 09, 2013 at 02:25:52PM -0700, Greg KH wrote: > > On Tue, Oct 08, 2013 at 03:32:21PM -0700, Kent Overstreet wrote: > > > On Wed, Oct 09, 2013 at 12:13:56AM +0200, Gabriel de Perthuis wro= te: > > > > Le 06/10/2013 17:50, Gabriel de Perthuis a =E9crit : > > > > > Le 06/10/2013 12:38, Cyril B. a =E9crit : > > > > >> Hello, > > > > >> > > > > >> I get the following oops immediately after booting on 3.10.1= 5. > > > > >> Everything works fine in 3.10.10. Both the backing and cache= devices are > > > > >> on top of mdadm. > > > > >=20 > > > > > Reverting c0f04d88e46d14de51f4baebb6efafb7d59e9f96 fixes it; = it was one > > > > > of the few commits that's in 3.12 and -stable but not in > > > > > bcache-for-3.11. That commit causes bch_insert_data to be ca= lled with > > > > > an unset op.cache_bio. > > > >=20 > > > > Pinging stable, > > > > http://git.kernel.org/linus/c0f04d88e46d14de51f4baebb6efafb7d59= e9f96 > > > > should be reverted. It was in 3.11.4 and 3.10.15 (and 3.12-rc3= ). > > > > It breaks bcache's writeback mode by calling bch_writeback_data= without > > > > setting cache_bio (it's more visible here[1]). There's a bit o= f > > > > indirection through closure calls, but it could never work. > > >=20 > > > Greg - I screwed up on that patch (and I'm still scratching my he= ad as > > > to how testing missed it) - it's a two line fix I'll be mailing o= ff to > > > Linus shortly, or if you don't want to wait for that reverting it= might > > > be better (though there was a data corruption bug that users repo= rted > > > that fixed, so I'd be hesitent there): > > >=20 > > > Here's the fix I just mailed out, just waiting on a bit of outsid= e > > > testing to mail it to Linus: > > >=20 > > > http://permalink.gmane.org/gmane.linux.kernel.bcache.devel/2113 > >=20 > > I'll wait for the new patch to get into Linus's tree, so as to keep > > things synced up. Let me know what the git commit it of the patch = is > > when hit hits his tree and I'll add it to the next stable queue. >=20 > It's in - it's 2fe80d3bbf1c8bd9efc5b8154207c8dd104e7306. Thanks! Now applied, thanks for pointing it out. greg k-h